OpenAI has introduced ChatGPT Dots, a new kind of AI agent that can carry out multi-step tasks across connected apps with limited input from the user. The company announced the feature on Tuesday, 10 November, according to a report by Gadgets360.

Dots are designed to work in the background. A user gives a task, the agent gets on with it across the apps it has been given access to, and it returns with an update or asks for approval when a decision is needed. The idea is that the user does not have to watch every step.

More than 4,000 apps are supported, according to the report. That range covers the tools people use for software development, research, sales and content creation, which are among the use cases OpenAI highlights for the feature.

Each agent gets its own separate cloud computer and browser, the report says. That separation means a Dot can work on a task independently, opening pages and using apps on the user's behalf, without taking over the user's own screen.

How users stay in control

Control is a central part of the design. Users decide which apps a Dot can access and what permissions it has, so an agent only reaches the tools and data the user has explicitly allowed.

Users can also connect multiple local computers to the same dot agent and choose which one it uses for each task, according to a report by ETV Bharat published on 10 November at 14:18 IST. That flexibility matters for people who work across a home computer and a work machine.

When a Dot reaches a step that needs a human decision, it pauses and asks for approval rather than guessing. That check is meant to keep the agent from taking irreversible actions, such as sending a message or making a purchase, without the user agreeing.

What Dots are for

OpenAI is positioning Dots as helpers for work that involves several steps across several tools. Research is one example: a Dot could gather material from different sources, organise it and present a summary. Software development, sales follow-up and content creation are others named in the announcement.

Specialised Dots for business workflows are being tested, according to the report, with wider availability planned. That suggests OpenAI sees the feature first as a productivity tool for teams, before it becomes a general-purpose assistant for everyday users.

The launch comes as rival AI companies race to move beyond chat. The next step for chatbots is widely seen as agents that can act, not just answer, and Dots are OpenAI's entry into that contest.

Availability details, including which ChatGPT plans get Dots first and when the feature reaches users in India, have not been fully set out in the reports so far. Users should check the official ChatGPT announcements for the rollout schedule for their plan and region.

As with any agent that can act across apps, the practical advice is to start with narrow permissions. Give a Dot access only to the apps a task needs, and review what it did before widening its reach.